984

Abstract

Introduction: Subcutaneous insulin may be utilized for the treatment of hyperglycemia in the intensive care unit (ICU) in selected patients; however, available evidence does not provide an optimal glucose management strategy to guide subcutaneous insulin product selection and dose titration. Methods: This was a prospective, open-label, randomized study. Patients were eligible for inclusion if they required initiation of basal insulin therapy in the medical ICU. They were randomized by bed number to receive either insulin glargine or NPH insulin. All other decisions regarding insulin management, including insulin dose titration, were made according to standard care. The primary outcome was hyperglycemic events, defined as blood glucose greater than 180 mg/dL. Secondary outcomes included hypoglycemia, defined as blood glucose less than 70 mg/dL; ICU length of stay (LOS); hospital LOS; and hospital mortality. Glucose data was clustered by patient and analyzed using a generalized estimating equation procedure. Results: From October 2012 to March 2013, a total of 55 patients were enrolled in the study. The total daily basal insulin dose was similar between the NPH insulin group (59 ± 47 units) and the insulin glargine group (72 ± 51 units, p = 0.34). Patients in the NPH insulin group were at a significantly lower risk for hyperglycemia when compared to patients in the insulin glargine group (RR 0.65, 95% CI 0.44 - 0.95). There was no difference between the two groups in the risk of hypoglycemia (RR 1.56, 95% CI 0.29 - 8.51). There were no significant differences found between the groups regarding ICU LOS, hospital LOS, or hospital mortality. Conclusions: NPH insulin was more effective than insulin glargine for reducing hyperglycemia in critically ill medical patients. Furthermore, there were no differences found in terms of the risk of hypoglycemia, ICU or hospital LOS, or hospital mortality.